Energy Efficient 3D Memristor-based Neuromorphic IC with Programmable Attention Mechanism

Speaker:  Hongyu An

Host: MICS

Date: November 10 (Friday), 2017
Time: 2:30 PM - 3:30 PM
Location: Whittemore 654 (6th Floor Conference Room)

Abstract:

Inspired by the neural architecture of the human brain, neuromorphic computing has been proposed to accommodate the ever-increasing demand for application variety. As a critical functionality of the perception system, attention mechanism enables the brain to devote more computing effort to process desired information. In this seminar, I will introduce our current research work on revealing the fundamentals behind this mechanism. The mathematical model of attention mechanism and its hardware implementations with nanoscale device memristor and three-dimensional (3D) monolithic integration technology will be presented. The vision disorder perception phenomena “Figure-ground alternation recognition” will also be discussed.

Bio:

Hongyu An is a Ph. D. student at the Bradley Department of Electrical and Computer Engineering (ECE) at the Virginia Tech. He received his B.S degree in Electrical Engineering from Shenyang University of Technology, Shenyang, China and M.S degree in Electrical Engineering from Missouri University of Science and Technology, Rolla, USA. During his Ph. D. study, he has published 3 journal articles and 9 international peer reviewed conference papers. One of his paper has been nominated as Best Paper Award in 2017 IEEE International Symposium on Quality Electronic Design (ISQED). He is also the recipient of the NSF Student Travel Fellowship Award at IEEE NANO 2017. His areas of interest include Neuromorphic Computing, Three-Dimensional Integrated Circuit (3D-IC) Design, and Artificial Intelligence.